Research News
New Model Shows Small NY Mobility Hubs Are Cutting Car Trips and Boosting Transit, Even With Sparse Data
Two pilot mobility hubs in New York's Capital District are nudging commuters out of their cars and onto buses, bikes, and car-share vehicles, according to a new analysis from NYU Tandon's C2SMART transportation research center.
But the same model suggests these demonstration hubs aren't necessarily located where they’d do the most good.
The hubs, one near UAlbany's downtown campus and the other in downtown Cohoes, work like transfer points, providing a bus stop paired with a shared bike dock that lets riders switch modes mid-trip instead of driving door to door. The Cohoes hub also offers car-share parking, a service not available at the UAlbany site.
The Tandon researchers estimate the hubs are cutting more than 75 vehicle miles driven each day and about 11 metric tons of carbon a year, roughly what two households produce annually. Riders pair two modes per trip, often combining a bus, bike-share, or walking leg with another mode, rather than driving or carpooling the whole way.
The hubs generate more than $5,000 a day in added value across both sites, a way of measuring how much better off travelers are with the extra option available, even if they never use it themselves, according to the researchers. It's not money changing hands, but a dollar figure economists use to capture things like time saved or an easier trip.
The researchers who led the study, C2SMART Deputy Director Joseph Chow, an NYU Tandon Institute Associate Professor, and Xiyuan Ren, a C2SMART postdoctoral fellow, published their findings in Transportation Research Part A.
Chow and Ren started with an existing model that predicts how New Yorkers get around, built from more than 50 million simulated trips across the state. That model can guess whether someone will drive, take transit, bike, or walk for a given trip. Mobility hubs were not reflected in the simulation because they did not exist when the underlying data was put together.
The researchers then taught the model something new: that hub trips, like driving to a bus stop then biking the rest of the way, are also an option. To make that addition realistic rather than a guess, they tuned it using real information, a 40-response survey of hub users and actual ridership numbers from the Capital District Transportation Authority (CDTA), the region's public transit agency.
They then reran that tuning process a thousand times on slightly different versions of the same small dataset, a standard statistical check to make sure their results weren't just a fluke of having so little real-world data to work with.
"You don't have enough real-world data to evaluate a pilot like this, but you also can't wait years to find out if it's working," Chow said. "Our approach borrows strength from a much larger travel model and uses the small amount of real data we have to calibrate it."
The team used the model to estimate outcomes at 1,100 locations across the region, using existing CDTA bus stops as hypothetical hub sites, not real pilots. Among all 1,100, the two actual hubs CDTA built, as part of a demonstration project that ran from April 2022 to June 2024, ranked only in the middle, roughly the 20th to 40th percentile, on measures like how many drivers they'd pull off the road and how much value riders would get.
The model suggests, in other words, that the two pilot sites weren't necessarily where a hub would have the biggest impact, likely because they were chosen for practical reasons, such as available land or existing partnerships, rather than for maximum benefit.
"It was a little surprising the two hubs CDTA built landed in the middle of the pack," Ren said. "That's not a criticism; pilots get chosen for practical reasons. But it suggests real value in running this kind of analysis before committing to a location."
The authors say the same approach, pairing a small on-site sample with a large behavioral model, could extend to other emerging services, such as autonomous shuttles or microtransit, before enough ridership data accumulates to build a model from scratch. Chow said the underlying statewide model isn't specific to mobility hubs, the team has previously used it to evaluate other mobility service programs, and since it draws on data available nationwide, the same approach could extend beyond New York.
Support for the research was provided by the New York State Energy Research and Development Authority (NYSERDA).
Xiyuan Ren, Joseph Y.J. Chow, A data fusion approach for mobility hub impact assessment and location selection: Integrating hub usage data into a large-scale mode choice model, Transportation Research Part A: Policy and Practice, Volume 211, 2026
New Study Highlights the Weather-Driven Costs of 24/7 Clean Power Matching
As demand for electricity surges from artificial intelligence, data centers, and other new technologies, companies are increasingly pledging to procure carbon-free electricity to mitigate emissions associated with their electricity use. While quantifying system-wide emissions is relatively straightforward, the complex nature of power grid operations makes it more challenging to ascertain the emissions impacts of a single consumer that also procures its own clean energy. This challenge is further compounded by clean energy sources like solar and wind, whose hour to hour generation can vary drastically, depending on weather patterns.
A new study by researchers at NYU Tandon and the Massachusetts Institute of Technology evaluated the system impacts of so-called 24/7 or hourly matching, involving matching grid electricity consumption with generation from procured clean electricity, a strategy of growing corporate and regulatory interest. By incorporating multiple years of weather variability into sophisticated power-system models, the researchers found that procuring clean electricity to match consumption every hour of the day can deliver emissions benefits, but at higher cost than previous studies suggested. The work also indicates that in regions already pursuing aggressive renewable energy policies, less strict approaches may achieve nearly the same climate benefits for considerably less money.
The study, published in Environmental Science & Technology, focuses on electricity-based production of hydrogen, a pathway of growing interest to decarbonize difficult-to-electrify segments of the economy like heavy industry. Recently, governments in both the United States and Europe have proposed or implemented rules requiring producers to match their electricity use with generation from newly built renewable energy resources, initially on an annual basis and eventually on an hourly basis. Beyond hydrogen, the findings apply broadly to other large electricity consumers, including data centers, and speak directly to ongoing revisions of the Greenhouse Gas Protocol's Scope 2 rules, the standard many companies use to account for and report the emissions tied to their electricity use, where hourly matching is a central proposal under debate.
Until now, however, most analyses on hourly matching have relied on a single year of weather data to characterize wind and solar resource variability. In reality, wind and solar resources fluctuate significantly from year to year.
"We wanted to understand how real-world, inter-annual weather variability changes the economics and emissions of clean electricity procurement," said senior author Dharik S. Mallapragada, associate professor of chemical and biomolecular engineering at NYU Tandon. "When projects are expected to operate for decades, designing them around a single weather year can paint an overly optimistic picture of both cost and performance."
To explore that question, the researchers modeled a grid-connected hydrogen facility operating in the Texas power grid, one of the nation's largest renewable energy markets. They optimized investments across nine representative years of historical wind and solar conditions, both individually and collectively, and then tested those designs for their robustness against additional weather scenarios.
The team's model consistently showed that hourly matching costs more than annual matching and that the cost premium is sensitive to the weather year, varying between ranging from $0.68 to $1.18/kg of H2 produced. For context, the average cost of fossil H2 production is around $1/kg. The team also showed that such single weather-year based investment plans can lead to shortfalls in clean electricity supply across multiple hours of the year when tested against additional weather scenarios not included in the planning analysis.
The cost of clean electricity procurement to be robust to inter-annual weather variations was found to lead to a still higher cost premium, of $1.29 per kilogram compared with annual matching. This cost premium stems from oversizing the capacity of wind and solar farms to manage shortfalls in supply as well as installing larger electrolyzers capable of ramping production up and down, along with extensive hydrogen storage to buffer periods when renewable generation falls short.
The researchers also examined ways to reduce costs of 24/7 matching without sacrificing most of the environmental benefits. Allowing hydrogen producers to match80 to 90 percent of their hourly electricity needs with clean power lowered costs while retaining much of the emissions benefits achieved under full compliance. Likewise, if hydrogen production operates within electricity grids already governed by strong renewable portfolio standards, annual matching performed nearly as well as hourly matching from an emissions standpoint, but at much lower costs. A further option, with similar emissions outcomes at a comparable or lower cost, is to fold new electricity demand directly into those existing renewable portfolio standards rather than layering separate, granular, matching requirements on top. These findings speak directly to ongoing debates among governments and corporations over the role of "24/7 matching" in voluntary and regulatory emissions reporting.
Michael A. Giovanniello, Dharik S. Mallapragada; Emissions and Cost Trade-Offs of Time-Matched Clean Electricity Procurement under Interannual Weather Variability: A Case Study of Hydrogen Production. Environ. Sci. Technol. 21 July 2026; 60 (28): 19854–19865. https://doi.org/10.1021/acs.est.6c00988
Inside the Urban Machine: Where America's Data Centers Actually Live
Updated August 3, 2026 to include Note on Methodology and Terminology.
When people picture a data center, they often imagine something remote: a huge warehouse humming quietly far from the city. New research from NYU Tandon School of Engineering shows that assumption is largely wrong.
The study, published in Nature Cities, examined the locations of 4,283 data centers across the contiguous United States and found that 97.5% of them sit inside metropolitan or micropolitan statistical areas, meaning urban cores and their immediate surroundings.
The roughly 2.5% of facilities technically outside city limits are, on average, just 8.5 miles from the nearest urban edge. The cloud lives downtown.
"There is a prevailing narrative of these data centers being somewhere in the middle of nowhere, in rural areas, being a positive force for employment, and being the future of rural communities," said lead researcher, NYU Tandon Institute Professor Maurizio Porfiri, who is the Director of Tandon’s Center for Urban Science + Progress and of the NYU Urban Institute. "We dramatically challenged this view."
The concentration is striking even within the urban category. Five metro areas, Washington-Arlington-Alexandria, Chicago, Dallas-Fort Worth, New York-Newark-Jersey City, and Phoenix, account for nearly a third of all U.S. facilities. The Washington region alone hosts 610 data centers, reflecting Northern Virginia's status as the global capital of data infrastructure (see appendix).
So why cities?
The answer, the researchers found, suggests that a lot has to do with what’s already there and what used to be there.
The single strongest predictor is electricity capacity, meaning how much power local generators can produce. Data centers are power-hungry, running thousands of servers around the clock and drawing enormous, steady loads from the grid.
But a notable finding goes beyond electricity supply. Closed coal plants near cities are becoming magnets for new data center builds. When a plant shuts down, the power lines and grid connections built to carry electricity continuously do not disappear. Data center developers can tap directly into that infrastructure, or, in some cases, redevelop the sites themselves.
The numbers bear this out. Among cities that overlap with areas designated as Energy Communities under a 2022 federal policy, data centers under development are twice as likely to be found there than in cities without that designation (see appendix).
The policy was designed to direct clean-energy investment toward regions hurt by coal plant closures. The research suggests the digital economy may be taking root in many of the same places as the fossil fuel economy it is meant to succeed.
Because data centers draw from their local grid, their carbon footprint depends heavily on how that grid generates power. A typical data center in Montana or North Dakota produces more than 350,000 tons of CO2 emissions annually, while the average facility in Vermont, New Hampshire, or Arkansas produces less than 3,000 tons.
Beyond electricity supply, data centers also cluster where IT workers and high-speed internet are concentrated. Local water shortages seem to have less of an impact on data center placement, despite the facilities consuming enormous amounts of water for cooling.
The opacity of the industry compounds all of these problems.
“As this industry rapidly grows, the limited publicly available data on its footprint creates a real challenge,' said Ofek Lauber Bonomo, a postdoctoral researcher in Porfiri’s Dynamical Systems lab and a paper co-author. “That makes it more difficult for planners, for local residents, and for anyone trying to make informed decisions about their community's future."
"The patterns we found were consistent and clear,” added Anton Rozhkov, a CUSP Industry Assistant Professor and paper co-author. “Where the infrastructure already exists, the data centers follow. The question now is whether that is the future we want to build.”
The Nature Cities paper follows the recent announcement that Porfiri and Camilla Ancona – a paper co-author and postdoctoral researcher in Porfiri’s Dynamical Systems Lab – were named among the 2026 cohort of Microsoft Research Fellows, to advance their work using AI-driven simulations to help utilities and regulators decide where to site data centers before breaking ground.
The research in the Nature Cities paper was supported by the NYU Abu Dhabi (NYUAD) Center for Interacting Urban Networks, funded by the Abu Dhabi government through the NYUAD Research Institute.
Ancona, C., Lauber Bonomo, O., Rozhkov, A. et al. Urban infrastructure and fossil fuel industrial legacy drive US data center siting. Nat Cities (2026). https://doi.org/10.1038/s44284-026-00487-z
Note on Methodology and Terminology:
This study analyzed a 2025 dataset of 4,283 commercial data centers in the contiguous United States from the commercial Data Center Map database. The dataset includes operational facilities as well as projects that were planned, under construction, or land-banked at the time of analysis. Throughout this article, "city" refers to metropolitan and micropolitan statistical areas (MSAs and MicroSAs), as defined by the U.S. Office of Management and Budget, which classify counties based on economic integration with an urban core rather than population density or land use. As a result, some counties classified as part of metropolitan or micropolitan areas may have low population density or predominantly rural land use. The study characterizes the overall geographic distribution of U.S. data centers by facility count during the study period.
Appendix
Data centers in each metro area - top 10
|
# |
Metro area |
Facilities |
Share of U.S. total |
|---|---|---|---|
|
1 |
Washington–Arlington–Alexandria |
610 |
14.2% |
|
2 |
Chicago–Naperville–Elgin |
241 |
5.6% |
|
3 |
Dallas–Fort Worth–Arlington |
192 |
4.5% |
|
4 |
New York–Newark–Jersey City |
163 |
3.8% |
|
5 |
Phoenix–Mesa–Chandler |
154 |
3.6% |
|
6 |
Atlanta–Sandy Springs–Roswell |
136 |
3.2% |
|
7 |
Columbus |
133 |
3.1% |
|
8 |
San Jose–Sunnyvale–Santa Clara |
130 |
3.0% |
|
9 |
Los Angeles–Long Beach–Anaheim |
90 |
2.1% |
|
10 |
Des Moines-West Des Moines |
76 |
1.8% |
Data centers in Energy Community (EC) designated areas - top 8
|
# |
Metro area |
Under-development data centers |
|---|---|---|
|
1 |
Chicago–Naperville–Elgin, IL–IN |
108 |
|
2 |
Dallas–Fort Worth–Arlington, TX |
58 |
|
3 |
Washington–Arlington–Alexandria, DC–VA–MD–WV |
26 |
|
4 |
San Antonio–New Braunfels, TX |
24 |
|
5 |
Reno, NV |
18 |
|
6 |
New Haven, CT |
12 |
|
7 |
Scranton–Wilkes-Barre, PA |
11 |
|
8 |
Monroe, LA |
10 |
Solar Panels Promise To Help Save the Environment. But What Happens When They Die?
Solar panels have become the most iconic symbol of the clean energy transition. They cover rooftops, stretch across deserts, and quietly convert sunlight into electricity without emitting carbon dioxide. But hidden behind this success story is a growing environmental challenge that few people think about: what happens when millions of those panels reach the end of their lives?
By 2050, discarded photovoltaic modules could generate more than 80 million metric tons of waste. Unless new recycling methods are developed, many of those panels could end up in landfills, taking with them valuable metals that required enormous amounts of energy and mining to produce in the first place.
"Solar panels are the clean energy infrastructure of the future," says Juanita Hidalgo, Assistant Professor of Chemical and Biomolecular Engineering at NYU Tandon. "But to make solar truly sustainable, we also need to think about what happens after these technologies reach the end of their lifetime."
A new perspective paper, coauthored by Hidalgo and post-doctoral researcher Sara Hamilton, argues that one of the most promising solutions lies not in hotter furnaces or more intensive manufacturing, but in chemistry. Researchers are increasingly turning to hydrometallurgy, a family of recycling techniques that uses liquids to selectively dissolve and recover valuable metals. Instead of heating an entire solar panel at temperatures approaching 2,000 degrees Celsius, hydrometallurgy carefully separates individual components at low temperatures so they can be reused in new devices.
"It's a much more selective approach," explains Hamilton. "Rather than treating a solar panel as waste, we're treating it as a source of valuable materials that can be recovered and put back into the supply chain."
Today's recycling systems recover relatively simple materials such as aluminum frames, glass, and copper wiring. But the heart of every solar panel contains metals that are both economically valuable and strategically important. Silver, indium, gallium, tellurium, and lead all play critical roles in different kinds of solar cells, yet recovering them remains technically difficult and often too expensive to justify. Current recycling methods frequently rely on pyrometallurgy, which uses extremely high temperatures to melt materials apart. While effective, the process consumes large amounts of energy and can make it difficult to separate individual metals cleanly.
Hydrometallurgy offers a more tunable alternative. Carefully selected solvents dissolve specific metals, which can then be purified and recovered for reuse. In principle, the approach requires less energy and can recover materials with much greater precision.
But the researchers found that not all solar technologies are equally easy to recycle. Conventional crystalline silicon panels, which account for roughly 95 percent of the global solar market, typically require strong acids such as nitric acid to extract valuable silver. Those acids work well, but they are corrosive, hazardous to handle, and difficult to recycle themselves. Thin-film solar cells face similar challenges. Although they contain smaller amounts of material overall, they rely on critical metals such as indium, gallium, and tellurium that are usually recovered using equally aggressive chemical treatments.
One surprise was the recyclability of the hottest new solar technology. Perovskite solar cells have generated enormous excitement because they can be manufactured at lower cost than conventional silicon while achieving record-setting efficiencies in the laboratory. The new perspective suggests they are also remarkably well suited for environmentally friendly recycling.
Unlike conventional solar cells, perovskites are built from layers connected by relatively weak chemical interactions. As a result, several recent studies have shown that researchers can recover one of their most important ingredients — lead — using something unexpectedly simple: hot water. As the water cools, the dissolved lead crystallizes back into a compound that can be used to manufacture new perovskite solar cells.
The authors argue that recyclability should become a design goal rather than an afterthought. Instead of maximizing efficiency first and worrying about disposal decades later, engineers could build future solar cells with disassembly and material recovery in mind from the very beginning.
When Disaster Strikes, People Often Flee to Places That Feel Familiar
When the Marshall Fire tore through suburban Colorado in late 2021, residents had only hours to decide where to go. Some fled to nearby towns. Others stayed farther away for weeks or months. Now a recent study published in Humanities and Social Sciences Communications suggests those decisions were shaped not only by distance or danger, but also by something more human: the pull of familiar communities and social ties.
Researchers at NYU Tandon analyzed anonymized mobile phone location data from more than 200,000 devices in Colorado before and after the fast-moving wildfire, which destroyed more than 1,000 homes and displaced thousands of people. They combined those movement patterns with demographic data and measures of social connectedness between neighborhoods. Their conclusion: evacuees were more likely to choose destinations that resembled their home communities or where they had stronger social links.
“Even during a chaotic emergency, people do not move randomly,” says lead author Takahiro Yabe, Assistant Professor of Technology Management and Innovation and the Center for Urban Science + Progress. “They tend to seek places where they feel socially connected or where the community feels familiar.”
The study adds nuance to how scientists understand evacuation behavior. Traditional models often assume people head to the nearest available safe place or to larger population centers. But this research found that social factors strongly influenced where people actually went.
Most evacuees relocated between 20 and 60 kilometers from the fire zone, suggesting many wanted to remain relatively close to home. Yet when researchers compared real evacuation destinations with simulated destinations based only on population size and distance, the real destinations scored significantly higher for demographic similarity and friendship connections. In other words, people often chose places where they knew someone, or places that looked socially like where they came from.
The findings also revealed inequality in who could access those familiar refuges. Residents from whiter, wealthier, and more highly educated neighborhoods were more likely to evacuate to destinations with stronger social similarity and connectedness. Black, Asian, and lower-income populations were less likely to do so. That gap may matter because social networks can provide practical help during crises: a spare bedroom, child care, transportation, local knowledge, or emotional support.
“Access to social capital can shape recovery just as much as physical damage does,” says Vaidehi Raipat, a PhD candidate and lead author on the paper. “If some groups have fewer options to relocate into supportive communities, that can deepen existing inequalities after disasters.”
The team also examined what happened after the initial evacuation. People who relocated to areas with stronger social connectedness were more likely to return home over the following months. But those who moved to places that were demographically similar to their original communities were somewhat less likely to return, suggesting that a comfortable temporary destination may sometimes become a longer-term alternative.
That distinction could help officials plan for future climate disasters, which are becoming more frequent and more destructive. Wildfires, floods, and storms increasingly force sudden movement, yet emergency planning still tends to focus on roads, shelters, and hazard maps rather than the social geography of where people want to go.
The researchers argue that disaster response could improve by accounting for community ties. Knowing where evacuees are likely to head could help agencies position aid, anticipate population surges, and better support displaced residents. It could also identify people who lack strong networks and may need more assistance.
The study focused on one wildfire, so its authors caution that patterns may differ in hurricanes, floods, or other disasters. Still, the broader message is clear: in moments of upheaval, people often search not just for safety, but for belonging.
Could Physics Replace the Computer Keeping Your Robot Upright?
A new discovery in physics could help engineers stabilize robots and structures without relying on complex sensors and control systems, and design metamaterials and network systems that are presently beyond reach.
The finding, published in Nature Communications by researchers at NYU Tandon School of Engineering and Stony Brook University, shows that a mechanical system can be kept stable simply by switching between two behaviors at the right rhythm, even when neither behavior is stable on its own. No sensors watching the motion. No software constantly correcting it.
Once the timing is set, the physics does the rest.
Many machines must constantly stabilize their motion — keeping a walking robot from tipping over or preventing an aircraft wing from vibrating uncontrollably. Robots and other actively controlled systems typically do this by monitoring their environment and correcting their motion in real time, which requires sensors, processing power and software.
To test an alternative, the researchers built what they informally call the Frankenstein oscillator: a thin plastic strip fixed at one end with a small weight at the tip, subject to multiple loading conditions.
They then created two different kinds of instability. A magnetic coil pushed the beam away from its resting position in a way similar to a ball balanced on a horse’s saddle: if it moves slightly off center, it slides away in certain directions. A small fan blew air across the strip, feeding energy into the motion so that its swings grew larger rather than fading away, similar to how a playground swing rises higher when someone pushes at the right moment.
Both forces were switched on and off in carefully timed pulses.
The result was striking. Stability appeared only within a narrow band of switching speeds, with periods between roughly 218 and 238 milliseconds. Inside that window the beam stayed nearly still. Outside it, the motion quickly grew and the beam swung away.
Why should switching between two unstable behaviors make anything stable?
The idea builds on Kapitza's pendulum, named after Russian Nobel laureate Pyotr Kapitza. Vibrate the base of an inverted pendulum at exactly the right frequency and it stays upright, with no one watching or adjusting. Unlike a person balancing a stick on one hand, constantly shifting to stop it from falling, Kapitza's pendulum requires no such attention. Physics takes over.
In the classic case, the vibration provides a stabilizing effect, making the system virtually alternate between one stable and one unstable state. The new research asked a different question: what would happen if there were no stable states at all, if the physics were always pushing the system away from its resting position?
The answer depends on the type of instability involved. The “sliding” type — like the ball on the saddle — has one special direction in which motion actually shrinks instead of growing. The “swinging” type continually rotates the motion through different directions.
If the switching is timed correctly, that rotation can steer the motion into the shrinking direction before it has time to run away. The two instabilities, surprisingly, end up stabilizing each other.
“I have been thinking about the problem of stabilization of unstable systems through switching for over two decades,” said the paper’s senior author Maurizio Porfiri, an NYU Tandon Institute Professor and Director of both the NYU Urban Institute and the Center for Urban Science + Progress (CUSP). “In between ups and downs on the research, I was almost convinced that stabilization of two unstable systems would require some form of nonlinearity or even chaotic dynamics, but that is not the case: a simple, linear mechanical system can do the trick. The solution was in front of me for years, an extension of the marvelous ideas presented by Landau and Lifshitz in their Mechanics textbook that my uncle gave to me as a gift when I took my undergraduate dynamics class."
The researchers developed the theory first and then confirmed it experimentally with the beam. The narrow stability window they observed in the lab closely matched what the mathematical model predicted.
"Honestly, I had no belief that we would be able to demonstrate this phenomenon experimentally, as this involved working with a system that not only is unstable, but also features multiple sources of instability,” said Paolo Celli, Assistant Professor in Civil Engineering at Stony Brook University and co-corresponding author of the study. “The joy we felt when our carefully-designed experiment showed that narrow stability window is hard to explain. I am now super excited to see how this dynamic stabilization idea can be applied to other structural and robotic systems on the verge of instability"
The broader implication is a new design philosophy. Instead of always trying to eliminate instability, engineers may sometimes be able to build stable systems out of unstable pieces, harnessing the laws of physics rather than fighting them
The research was supported by the National Science Foundation through grants to both institutions. Along with Porfiri — who wrote about this research in a Behind the Paper post — and Celli, David Xiedeng — a Ph.D. student in Celli’s lab — is a co-author on the paper.
Watching a Film Reveals How the Brain Balances Eyes and Ears
For most of us, watching a movie feels effortless. We follow dialogue, read facial expressions, notice music cues and shifting scenery, and somehow fuse it all into a coherent story. But beneath that smooth experience, the brain is constantly deciding which sensory stream matters most in each moment.
A new study suggests that the frontal cortex, a region associated with planning and higher cognition, may act as a kind of traffic controller for this process — dynamically shifting attention between what we hear and what we see as a story unfolds.
To investigate, neuroscientists recorded brain activity directly from 19 epilepsy patients who had temporarily implanted electrodes for clinical monitoring. While in the hospital, participants watched a 12-minute multilingual short film* containing scenes in English, Greek, German, and French. Some foreign-language scenes included English subtitles, creating a natural test of how the brain handles changing audiovisual demands. Because the electrodes sat on or inside the brain, the researchers could track neural responses with millisecond precision, far faster than MRI scans allow.
They found that the frontal cortex was not processing all sensory information equally. Instead, it showed a striking internal division. Ventral, or lower, frontal regions responded more strongly to auditory information, while dorsal, or upper, frontal regions were more tuned to visual input.
“This suggests the frontal cortex has an organized map for handling different kinds of information during real-world experiences,” said first author Faxin Zhou, a Ph.D. candidate in the NYU Tandon Biomedical Engineering Department. “It is not just a general control center, it appears to separate sound and sight in a structured way.”
The pattern became even more interesting when the language changed. During English-language scenes, when listeners could understand speech directly, frontal brain areas leaned more heavily on auditory processing. But during scenes in unfamiliar languages, activity shifted toward visual regions, suggesting viewers relied more on facial expressions, gestures, and subtitles to follow the plot.
To confirm that interpretation, the team recruited online volunteers to rate short clips from the film. Participants judged which moments were most important to understanding the story and whether audio or visual cues were more useful in each scene. Those ratings closely matched the neural data: spoken English favored sound, while foreign-language scenes favored visual cues. In other words, the brain appears to reweight its sensory priorities on the fly.
“When comprehension through speech becomes harder, the brain flexibly reallocates resources toward visual signals,” said senior author Adeen Flinker, Associate Professor of Biomedical Engineering at NYU Tandon and Neurology at NYU Grossman School of Medicine. “That adaptability may be essential for navigating everyday environments filled with competing information.”
The findings help illuminate a long-standing question in neuroscience: how the brain merges multiple senses in realistic settings. Much prior research has relied on simplified laboratory tasks. Movies, by contrast, more closely resemble real life, where sensory cues arrive continuously and unpredictably.
The study also hints that the frontal cortex may do more than merge information after the fact. It may actively decide which stream — sound or sight — deserves priority before conscious understanding emerges.
That insight could have practical implications. Better understanding how the brain reallocates sensory attention may help researchers design therapies for people with language disorders, autism, attention deficits, or hearing loss. It could also inspire more adaptive artificial intelligence systems that shift between audio and visual inputs depending on context.
The work has limitations. Because the participants were hospital patients with epilepsy, they may not perfectly represent the general population. Electrode placement was determined by medical need, not experimental design, leaving some brain areas less sampled than others. Still, the precision of direct neural recording offers a rare glimpse into how the living human brain manages everyday perception.
*"Foreign Language," Adam Kelly Morton, Ack, No Ledge Creative
Zhou, F., Khalilian-Gourtani, A., Dugan, P. et al. Frontal cortex organization supporting audiovisual processing during naturalistic viewing. Nat Commun 17, 5355 (2026). https://doi.org/10.1038/s41467-026-73947-8
Seeing Through a Robot’s Eyes: Augmented Reality Helps Humans Predict Machine Behavior
As robots increasingly move out of factories and into workplaces, hospitals, warehouses and public spaces, a simple challenge becomes increasingly important: helping people understand what those machines are about to do.
A new study by researchers at Bowling Green State University (BGSU) and NYU Tandon suggests that augmented reality (AR) may offer a surprisingly effective solution. By overlaying a robot's goals, planned routes and safety zones onto the real world through a smartphone, the researchers found that people became significantly better at anticipating robot behavior and identifying potential hazards.
The work, published in the journal Empathic Computing, addresses a growing concern in human-robot interaction: transparency. While robots are becoming increasingly autonomous, their decision-making processes often remain opaque to nearby humans. That uncertainty can create confusion, reduce trust and, in some situations, compromise safety.
"One of the biggest challenges in human-robot collaboration is helping people understand what a robot intends to do before it acts," says co-author Vikram Kapila, Professor of Mechanical and Aerospace Engineering at NYU Tandon. "When users can see a robot's planned path, destination and safety boundaries, they are better able to anticipate its actions and make informed decisions about their own movements."
To tackle the problem, the researchers developed a smartphone-based AR application that communicates a mobile robot's intentions in real time. Using a standard Android phone equipped with Google's ARCore software, the system overlays digital information directly onto the user's view of the physical environment.
The application provides three types of visual information. One mode displays the robot's destination as a virtual location pushpin. Another reveals the route the robot plans to follow. A third shows a digital twin of the robot itself moving through the environment, complete with a visual buffer zone indicating areas where collisions or interference could occur.
Unlike many previous AR systems, which rely on specialized headsets or projection hardware, the new approach works on an ordinary smartphone. That simplicity could make the technology easier to deploy in workplaces where workers and autonomous machines routinely share space.
To test whether the system actually improved understanding, the researchers recruited 58 participants with varying levels of experience in robotics and augmented reality. Participants viewed a series of AR scenarios showing robot navigation tasks and then answered questions designed to measure what human-factors researchers call situational awareness — the ability to perceive, understand and predict events unfolding in an environment.
The evaluation was based on the Situational Awareness Global Assessment Technique, or SAGAT, a widely used framework that measures three levels of awareness: perception of relevant information, comprehension of its meaning and projection of future events.
Participants were asked to identify robot goals, recognize obstacles, determine whether objects would interfere with the robot's movement and predict which areas of the environment would remain safe for human occupancy.
The results were striking. Across all tasks, participants achieved an average situational-awareness score of 86.5 percent. They were particularly successful at recognizing obstacles and identifying safe zones where they could avoid interfering with the robot's operation.
Just as important, participants reported feeling more confident about working alongside robots. More than 96 percent said the AR interface improved their understanding of robot intentions and increased their confidence in predicting robot behavior.
"The findings demonstrate that even a lightweight, smartphone-based AR system can substantially improve people's awareness of a robot's goals and movements," Kapila says. "That increased understanding is an important foundation for building trust, safety and effective collaboration between humans and autonomous systems," said paper’s lead author Sonia Chacko, Assistant Professor at BGSU, who received her doctoral degree from NYU Tandon.
The study arrives at a moment when robots are becoming more common in settings that were once the exclusive domain of humans. Warehouses increasingly rely on autonomous mobile robots to move goods. Hospitals are experimenting with robotic delivery systems. Service robots are beginning to appear in airports, hotels and retail environments. In such settings, the ability to quickly understand what a robot is planning to do may be as important as the robot's ability to understand human behavior.
Kapila also recently developed a mixed-reality system that allows people to communicate force instructions to a robot using a tablet. Instead of programming the robot through complex controls, users place a virtual arrow on the tablet screen over a real-world object. The arrow's position tells the robot where to apply force, its orientation indicates the direction of the force, and its length specifies how much force should be used. The robot then carries out the task and provides visual feedback through a moving virtual indicator that shows whether the desired force has been achieved.
The findings reported in the journal Machines, authored by Christian Lourido, Kishan Reddy Raghunath and Kapila, suggest that mixed reality can make human-robot collaboration more intuitive by allowing people to communicate complex physical intentions visually rather than through specialized programming or expensive haptic equipment. Such technology could eventually be useful in manufacturing, healthcare, and other environments where humans and robots must work together safely and precisely.
The findings suggest that making robotic intentions visible may help bridge one of the most persistent gaps in human-machine collaboration. As autonomous systems become more common in everyday life, a digital window into a robot's plans could make working alongside them feel far more comfortable.
Chacko S, Kapila V. Making robots understandable: Augmented reality for enhancing situational awareness in human–robot co-located environments. Empath Comput. 2026;2:202517. https://doi.org/10.70401/ec.2026.0016
Protected Bike Lanes Causally Increase NYC Bikeshare Ridership, But Benefits Are Not Distributed Equally
Protected bike lanes increase Citi Bike ridership in New York City, but painted bike lanes and sharrows do not show a statistically significant causal effect on ridership after accounting for confounding factors, according to a new study from researchers at NYU's Tandon School of Engineering published this week in npj Sustainable Mobility and Transport.
The findings address a longstanding question in transportation planning: if and to what extent do different types of bicycle infrastructure actually encourage more people to ride.
Protected bike lanes physically separate cyclists from vehicle traffic using barriers such as curbs, parked cars, or flexible posts. Painted bike lanes provide only a painted stripe between cyclists and cars, while sharrows are bicycle symbols painted onto shared traffic lanes.
Using approximately 72 million Citi Bike trips recorded between 2013 and 2024 (a period of significant ridership growth), the researchers linked trip data to bicycle infrastructure located near stations across New York City.
Initial results suggested that both protected and painted facilities were associated with increased ridership. Stations near newly-installed protected bike lanes saw an average increase in trips of 18%, while stations near painted bike lanes and sharrows experienced an average increase of about 14%.
However, those initial before-and-after comparisons do not account for the fact that bike lanes are often installed in areas where cycling activity is already increasing.To isolate the effects of the infrastructure itself, the researchers used propensity score matching and difference-in-differences analysis, statistical methods designed to compare similar locations while controlling for pre-existing neighborhood characteristics and ridership trends.
After applying those methods, only protected bike lanes showed a statistically significant causal effect on Citi Bike ridership. The researchers estimated an average increase of approximately 379 additional rides per station per month following installation of protected lanes. In contrast, painted bike lanes and sharrows did not show a statistically significant causal effect on ridership.
"Not all bike lanes are created equal," said Takahiro Yabe, Assistant Professor in the Department of Technology Management and Innovation (TMI) and the Center for Urban Science + Progress (CUSP) at NYU Tandon School of Engineering. "When cities invest in cycling infrastructure, the design details can determine whether a lane simply exists on a map or actually changes how people travel. That matters for transportation, public health, and sustainability, especially when cities are making difficult choices about how to invest limited resources."
"Painted bike lanes and sharrows may cost less and face less political pushback, but we now have evidence at a massive scale that protected bike lanes are really what can move the needle on ridership," said Marcel Moran, the lead author of the paper. Moran is currently an Assistant Professor at San José State University, and was a Faculty Fellow at CUSP during this project.
The study also examined whether the effects of protected bike lanes differed across neighborhoods. The researchers found that the positive ridership effect was statistically significant only in Census block groups with the lowest share of Black residents. In neighborhoods with higher shares of Black residents, they did not detect a statistically significant causal effect on Citi Bike ridership.
"Protected bike lanes seem to work best where cycling was already a realistic option for people,” said Malik Salman, a paper co-author. Salman is an NYU CUSP alumni and currently a CUSP Research Scholar in Yabe’s lab. “In communities where residents face other barriers — cost, discriminatory policing, a history of being left out of the planning process — the infrastructure alone may not be enough to change behavior. That's not an argument against building protected lanes. It's an argument for doing more alongside them."
The results were more encouraging for older adults. In Census block groups with the highest share of residents between ages 60 and 79, protected bike lanes produced particularly strong ridership gains. The researchers suggest that older adults may be especially responsive to infrastructure that reduces perceived traffic-safety risks. This aligns with evidence from cities like Copenhagen, which feature an expansive network of protected bike lanes, as well as high ridership among older adults.
The study comes as New York City's bicycle network has expanded from roughly 900 miles of bike lanes in 2014 to approximately 1,500 miles by 2024, while Citi Bike recorded a system-high of roughly 45 million trips in 2024. The authors say their methodology could be applied to other cities with publicly available bikeshare and bike-lane data, including Chicago, Boston, San Francisco, and Washington, D.C.
Moran, M., Salman, M. & Yabe, T. Heterogeneous impacts of protected bike lanes on bikeshare behavior across demographic groups in New York. npj. Sustain. Mobil. Transp. 3, 39 (2026). https://doi.org/10.1038/s44333-026-00107-2
A 3-D Printed Stent, Shaped Like a Lily, Could Speed Recovery After Weight-Loss Surgery
Each year, about 250,000 Americans undergo sleeve gastrectomy, one of the most common weight-loss operations in the United States. For most patients, recovery is uneventful.
But for a small share — between one and three percent in routine cases, and as many as one in ten in revision surgeries — the procedure can leave behind a gastric leak, in which fluid escapes from the stomach and forms an abscess.
Treating those leaks can be a long process. Doctors usually rely on endoscopic internal drainage, threading a small plastic tube called a double-pigtail stent through the stomach wall so the fluid can drain. But the devices typically used are built for bile ducts, not for the oddly shaped cavities created by gastric leaks.
That mismatch matters. The stents can slip, drain slowly and require repeated procedures before the leak resolves.
Now, researchers at New York University say they have found a better approach by changing the shape of the stent itself. Their findings have been published in Advanced Healthcare Materials.
Their prototype, called the Lily stent, is the first product of a design framework the researchers named PETALS, for Personalized Endoscopic Transmural Abscess Leak Solution, a mathematical approach to optimizing drain shape for complex biological fluids that the team says could be applied beyond gastric leaks to other drainage challenges in the body.
Using computer simulations and mathematical modeling, the researchers found that length and inner diameter mattered most, while the curled anchoring ends had little effect on fluid flow. Counterintuitively, a wider tube does not drain better. Increasing the inner diameter shrinks the gap around the outside of the stent, where most of the fluid actually travels. That means exterior topography, not interior volume, is the primary driver of drainage performance.
That insight is the foundation of the PETALS framework. By mathematically optimizing the outer surface geometry for the viscosity and pressure of gastric fluid, the team arrived at their new stent, a six-part structure that creates more effective routes for fluid to move around the device.
“The key insight is that the geometry of the tube’s cross-section, especially the exterior surface, fundamentally determines how fast fluid moves through and around it,” said Khalil Ramadi, an assistant professor at NYU Abu Dhabi and NYU Tandon School of Engineering and the study’s senior author. “We’re not just making it out of a different material. We’re changing the shape to make it work better.”
If the results hold up beyond the lab, the payoff could be meaningful. Roughly 2,500 people in the United States require treatment each year for gastric leaks after bariatric surgery. Faster drainage could shorten recovery and reduce the need for repeat procedures, easing both the burden on patients and the cost of care.
The device is still early in development. So far it has been tested only in simulations and benchtop models. Animal studies will be needed before it can move closer to clinical use.
The Lily stent also proved more flexible than the commercial polyethylene device it is meant to replace, an attribute surgeons associate with better patient tolerance and reduced tissue damage. In short-term animal studies, tissue surrounding the implanted material showed no significant difference from tissue around standard polyethylene, an early indicator of biocompatibility.
The researchers note that the Lily design's constant cross-section geometry would allow it to be manufactured by conventional extrusion methods, without requiring hospitals to invest in 3-D printing infrastructure.
"Our work shifts the focus from just placing a stent to engineering its function at a structural level," said Parima Phowarasoontorn, a research assistant in Ramadi’s NYU Abu Dhabi lab and the paper’s first author. "Instead of simple tubes, we introduce cross-sectional designs that improve drainage while remaining compatible with existing endoscopic delivery procedures."
The Lily stent research follows Ramadi's lab's announcement of the CORAL capsule, an ingestible pill whose coral-like structure traps bacteria from the small intestine, enabling researchers to study microbial communities that contribute to certain diseases. Both devices draw on nature's own geometry to solve medical problems that conventional tools have struggled to address.
Phowarasoontorn P, Ko Y, Barajas-Gamboa JS, Pantoja JP, Al-Ketan O, Ali M, Sohn S, Naser HT, Dabbour AH, Khlaifat B, AlZubaidi A, Vega CA, Rodriguez J, Kroh M, Ramadi KB. Enhanced Endoscopic Internal Drainage of Gastric Abscess Through Additively Manufactured Stents. Adv Healthc Mater. 2026 Apr 2:e05860. doi: 10.1002/adhm.202505860.